Inosine released after hypoxia activates hepatic glucose liberation through A(3) adenosine receptors

Abstract

Inosine, an endogenous nucleoside, has recently been shown to exert potent effects on the immune, neural, and cardiovascular systems. This work addresses modulation of intermediary metabolism by inosine through adenosine receptors (ARs) in isolated rat hepatocytes. We conducted an in silico search in the GenBank and complete genomic sequence databases for additional adenosine/inosine receptors and for a feasible physiological role of inosine in homeostasis. Inosine stimulated glycogenolysis (approximately 40%, EC50 4.2 x 10(-9) M), gluconeogenesis (approximately 40%, EC50 7.8 x 10(-9) M), and ureagenesis (approximately 130%, EC50 7.0 x 10(-8) M) compared with basal values; these effects were blunted by the selective A3 AR antagonist 9-chloro-2-(2-furanyl)-5-[(phenylacetyl)amino][1,2,4]-triazolo[1,5-c]quinazoline (MRS 1220) but not by selective A1, A2A, and A2B AR antagonists. In addition, MRS 1220 antagonized inosine-induced transient increase (40%) in cytosolic Ca2+ and enhanced (90%) glycogen phosphorylase activity. Inosine-induced Ca2+ mobilization was desensitized by adenosine; in a reciprocal manner, inosine desensitized adenosine action. Inosine decreased the cAMP pool in hepatocytes when A1, A2A, and A2B AR were blocked by a mixture of selective antagonists. Inosine-promoted metabolic changes were unrelated to cAMP decrease but were Ca2+ dependent because they were absent in hepatocytes incubated in EGTA- or BAPTA-AM-supplemented Ca2+-free medium. After in silico analysis, no additional cognate adenosine/inosine receptors were found in human, mouse, and rat. In both perfused rat liver and isolated hepatocytes, hypoxia/reoxygenation produced an increase in inosine, adenosine, and glucose release; these actions were quantitatively greater in perfused rat liver than in isolated cells. Moreover, all of these effects were impaired by the antagonist MRS 1220. On the basis of results obtained, known higher extracellular inosine levels under ischemic conditions, and inosine's higher sensitivity for stimulating hepatic gluconeogenesis, it is suggested that, after tissular ischemia, inosine contributes to the maintenance of homeostasis by releasing glucose from the liver through stimulation of A3 ARs.

Full text

doi:10.1152/ajpendo.00173.2005
290:940-951, 2006. First published Dec 13, 2005;Am J Physiol Endocrinol Metab
Rafael Villalobos-Molina and Enrique Piña
Raquel Guinzberg, Daniel Cortés, Antonio Díaz-Cruz, Héctor Riveros-Rosas,
You might find this additional information useful...
66 articles, 23 of which you can access free at: This article cites http://ajpendo.physiology.org/cgi/content/full/290/5/E940#BIBL
1 other HighWire hosted article: This article has been cited by
[PDF] [Full Text] [Abstract] , February 1, 2008; 294 (2): G401-G410. Am J Physiol Gastrointest Liver Physiol
V. L. Kolachala, R. Bajaj, M. Chalasani and S. V. Sitaraman
Purinergic receptors in gastrointestinal inflammation
including high-resolution figures, can be found at: Updated information and services http://ajpendo.physiology.org/cgi/content/full/290/5/E940
can be found at: AJP - Endocrinology and Metabolismabout Additional material and information http://www.the-aps.org/publications/ajpendo
This information is current as of March 10, 2009 .
http://www.the-aps.org/.
20814-3991. Copyright © 2005 by the American Physiological Society. ISSN: 0193-1849, ESSN: 1522-1555. Visit our website at
organization. It is published 12 times a year (monthly) by the American Physiological Society, 9650 Rockville Pike, Bethesda MD
publishes results of original studies about endocrine and metabolic systems on any level ofAJP - Endocrinology and Metabolism
on March 10, 2009 ajpendo.physiology.orgDownloaded from

Inosine released after hypoxia activates hepatic glucose liberation
through A
3
adenosine receptors
Raquel Guinzberg,
1
Daniel Corte´s,
1
Antonio Dı´az-Cruz,
2
He´ctor Riveros-Rosas,
1
Rafael Villalobos-Molina,
3
and Enrique Pin˜a
1
1
Departamentos de Bioquı´mica, Facultad de Medicina;
2
Nutricio´n Animal y Bioquı´mica,
Facultad de Medicina Veterinaria y Zootecnia; and
3
Unidad de Biomedicina, Facultad
de Estudios Superiores-Iztacala, Universidad Nacional Auto´noma de Me´xico, Mexico City, Mexico.
Submitted 20 April 2005; accepted in final form 3 December 2005
Guinzberg, Raquel, Daniel Corte´s, Antonio Dı´az-Cruz, He´ctor Ri-
veros-Rosas, Rafael Villalobos-Molina, and Enrique Pin˜a. Inosine re-
leased after hypoxia activates hepatic glucose liberation through A
3
adeno-
sine receptors. Am J Physiol Endocrinol Metab 290: E940 –E951, 2006. First
published December 13, 2005; doi:10.1152/ajpendo.00173.2005.—Inosine,
an endogenous nucleoside, has recently been shown to exert potent
effects on the immune, neural, and cardiovascular systems. This work
addresses modulation of intermediary metabolism by inosine through
adenosine receptors (ARs) in isolated rat hepatocytes. We conducted
an in silico search in the GenBank and complete genomic sequence
databases for additional adenosine/inosine receptors and for a feasible
physiological role of inosine in homeostasis. Inosine stimulated gly-
cogenolysis (⬇40%, EC
50
4.2 ⫻10
⫺9
M), gluconeogenesis (⬇40%,
EC
50
7.8 ⫻10
⫺9
M), and ureagenesis (⬇130%, EC
50
7.0 ⫻10
⫺8
M)
compared with basal values; these effects were blunted by the selec-
tive A
3
AR antagonist 9-chloro-2-(2-furanyl)-5-[(phenylac-
etyl)amino][1,2,4]-triazolo[1,5-c]quinazoline (MRS 1220) but not by
selective A
1
,A
2A
, and A
2B
AR antagonists. In addition, MRS 1220
antagonized inosine-induced transient increase (40%) in cytosolic
Ca
2⫹
and enhanced (90%) glycogen phosphorylase activity. Inosine-
induced Ca
2⫹
mobilization was desensitized by adenosine; in a
reciprocal manner, inosine desensitized adenosine action. Inosine
decreased the cAMP pool in hepatocytes when A
1
,A
2A
, and A
2B
AR
were blocked by a mixture of selective antagonists. Inosine-promoted
metabolic changes were unrelated to cAMP decrease but were Ca
2⫹
dependent because they were absent in hepatocytes incubated in
EGTA- or BAPTA-AM-supplemented Ca
2⫹
-free medium. After in
silico analysis, no additional cognate adenosine/inosine receptors
were found in human, mouse, and rat. In both perfused rat liver and
isolated hepatocytes, hypoxia/reoxygenation produced an increase in
inosine, adenosine, and glucose release; these actions were quantita-
tively greater in perfused rat liver than in isolated cells. Moreover, all
of these effects were impaired by the antagonist MRS 1220. On the
basis of results obtained, known higher extracellular inosine levels
under ischemic conditions, and inosine’s higher sensitivity for stim-
ulating hepatic gluconeogenesis, it is suggested that, after tissular
ischemia, inosine contributes to the maintainence of homeostasis by
releasing glucose from the liver through stimulation of A
3
ARs.
ischemia; calcium; urea; phylogenetic analysis; homeostasis.
INOSINE IS A NATURALLY OCCURRING PURINE NUCLEOSIDE formed by
adenosine deamination. Its normal interstitial concentrations in
rat plasma and serum have been reported in the range of 0.5–20
␮M (51, 61), and inosine accumulates to even higher levels
(⬎100 ␮M) than adenosine does in ischemic tissues (34, 41,
50, 51, 56). Our laboratory was the first to describe a stimu-
latory action of inosine on ureagenesis and gluconeogenesis in
isolated hepatocytes (23, 68). However, over the last decade
several reports (e.g., Refs. 19, 32, 59) appeared regarding the
role of inosine in regulating the immunologic and cardiovas-
cular systems. Although in the majority of cases inosine binds
to A
3
adenosine receptors (ARs) to promote its effects (19, 32,
59), there are reports in which A
2A
AR (19) or even an
AR-independent G protein-coupled receptor (GPCR) pathway
(27) were involved.
To date, four AR subtypes have been cloned (A
1
,A
2A
,A
2B
,
and A
3
), each with unique tissue distributions, ligand affinity,
and signal-transducing mechanism (for a review, see Ref. 49).
All four AR subtypes are present in isolated hepatocytes,
where they stimulate glycogenolysis, gluconeogenesis, and
ureagenesis rates (49). Signal transduction systems for obtain-
ing these increases were via adenylyl cyclase for A
2A
and A
2B
AR, whereas A
1
and A
3
AR involved changes in cytosolic
Ca
2⫹
(20 –22, 60, 66). The purpose of this work included the
following: 1) to define the receptor type involved in inosine
responses in isolated hepatocytes; 2) to identify the signal
transduction pathway mediating these inosine responses; 3)to
explore the possibility of finding additional adenosine/inosine
receptors; and 4) to obtain insight into the physiological mean-
ing of these inosine actions.
MATERIALS AND METHODS
Selective AR agonists and antagonists used in this work are
included in Table 1 and are listed in alphabetical order of their
abbreviations. Full chemical names, the receptor-binding constant for
AR agonist, reported data on the K
i
for AR antagonists, and pertinent
references are additionally included. All of these compounds were
purchased from Sigma RBI.
All animal experiments were conducted in accordance with the
Federal Guidelines for the Care and Use of Animals (NOM-062-
ZOO-1999, Ministry of Agriculture, Mexico) and were approved by
the Institutional Ethics Committee of the National Autonomous Uni-
versity of Mexico’s Faculty of Medicine (FM-UNAM).
Isolation of hepatocytes. Male Wistar rats (150 –200 g) were
anesthetized with ether, and cells were isolated by the method of
Berry and Friend (7) as modified by Guinzberg et al. (23). Hepato-
cytes were used when viability was at least 95%, as assayed by the
trypan blue exclusion method. Experiments were conducted by dupli-
cate with 30 – 40 mg wet wt hepatocytes.
Address for reprint requests and other correspondence: E. Pin˜ a, Departa-
mento de Bioquı´mica, Facultad de Medicina, Universidad Nacional Auto´noma
de Me´xico, Apdo. Postal 70159, Mexico City, 04510, Mexico (e-mail:
epgarza@servidor.unam.mx).
The costs of publication of this article were defrayed in part by the payment
of page charges. The article must therefore be hereby marked “advertisement”
in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
Am J Physiol Endocrinol Metab 290: E940–E951, 2006.
First published December 13, 2005; doi:10.1152/ajpendo.00173.2005.
0193-1849/06 $8.00 Copyright ©2006 the American Physiological Society http://www.ajpendo.orgE940
on March 10, 2009 ajpendo.physiology.orgDownloaded from

Ureagenesis. Hepatocytes from 24-h-starved rats were incubated
for 1 h at 37°C in an atmosphere of O
2
-CO
2
(95%-5%) for 60 min in
a gyratory water bath in Krebs-Ringer buffer (KRB) containing 10
mM glucose, 5 mM (NH
4
)
2
CO
3
, and 3 mM ornithine. Urea synthesis
was assayed after 60 min (24).
Gluconeogenesis. Hepatocytes from 24-h-starved rats were incu-
bated for1hinKRBcontaining 10 mM lactate. Glucose synthesis
was measured in the supernatant of cells by the glucose oxidase
method (18).
Glycogenolysis. Hepatocytes from rats fed ad libitum were incu-
bated for 45 min in KRB without lactate or any other substrate.
Glucose release was measured (18).
Glycogen phosphorylase activity. This activity was assayed by
measuring the incorporation of [U-
14
C]glucose 1-phosphate into gly-
cogen, as described by Starke et al. (57). Hepatocytes were exposed to
the agents, and aliquots were withdrawn at time intervals and placed
in 0.2 ml of ice-cold medium containing 10 mM MES, 20 mM NaF,
25 mM glycerophosphate, 10 mM EDTA, and 0.8 mM digitonin.
Hepatocyte extracts (25 ␮l) were mixed with an equal volume of
phosphorylase assay medium containing 50 mM NaF, 4.8 mM caf-
feine, 86 mM glucose 1-phosphate, 2% glycogen, and 8.5 ␮Ci of
[U-
14
C]glucose 1-phosphate and incubated at 37°C. The reaction was
stopped after 30 min by the addition of 25 ␮l of glacial acetic acid. A
50-␮l sample was spotted onto filter paper and washed twice with
66% ethanol, washed with acetone, and placed in a cocktail for liquid
scintillation counting.
cAMP accumulation. Hepatocytes from fed rats were incubated at
37°C for 2 min in KRB. cAMP was measured using the Amersham kit
TRK4312.
Ca
2
⫹
measurement in fura 2-AM loaded hepatocytes. This was
performed as described by Llopis et al. (42). Briefly, isolated hepa-
tocytes from fed rats were diluted in KRB to a final concentration of
40 mg wet wt/ml and incubated for 10 min at 37°C in an atmosphere
of O
2
-CO
2
(95%-5%). Cells were incubated for an additional 20 min
in the presence of 3 ␮M fura 2-AM and were washed twice by
centrifugation at 500 rpm for 3 min. Liver cells were divided into
200-␮l aliquots, immersed in ice, and used within the subsequent 5
min. Ca
2⫹
was measured in these cells as in Llopis et al. (42) using
aK
d
⫽224 nM.
Hypoxia/reoxygenation in isolated hepatocytes and perfused liver.
In experiments with fed rats, isolated hepatocytes were used to
measure inosine, adenosine, and glycogenolysis release. Fasted rats
(16 h) were used to measure gluconeogenesis and ureagenesis rates.
Rat livers were perfused in situ by placing a cannula in the portal vein,
and KRB was equilibrated with an O
2
-CO
2
mixture (19:1) at a
constant flow rate of 16 ml/min. Hepatic venous effluents were
obtained via a cannula in the vena cava.
Adenosine and inosine release quantification. Nucleosides were
measured by enzymatic assay in double-beam spectrophotometer by
the method described by Olsson (47).
Statistical methods. Values are reported as means ⫾SE. Student’s
t-test was applied to assess differences between groups. Statistical
significance was set at P⬍0.05.
Identification of cognate ARs on protein databases. Initially, se-
quences of known ARs from the rhodopsin superfamily were retrieved
from the Swiss-Prot protein database at http://au.expasy.org/sprot/ (3).
The amino acid sequence from each of these known ARs was used as
bait for BLASTP (1) searches at the National Center for Biotechnol-
ogy Information GenBank nonredundant protein database (6). To
determine the number of sequences encoding ARs in animals with
complete genome sequence, we repeated the BLAST search with the
tBLASTn program (1), using amino acid sequences of characterized
adenosine GPCRs as queries against whole genomic DNA sequences
or the high-throughput genomic sequence database from human,
mouse, rat, zebra fish, Japanese puffer fish (International Fugu Ge-
nome Consortium, assembly version 3.0; http://genome.jgi-psf.org/
fugu6/fugu6.home.html), and the ascidian Ciona intestinalis (assem-
bly version 1.0; http://genome.jgipsf.org/ciona4/ciona4.home.html).
Ab initio gene predictions were performed with the GeneComber
system (54), which provides increased gene recognition accuracy by
combining predictions from the gene-finding Genscan (10) and
HMMgene (37) programs. GeneComber-predicted exons were veri-
fied by multiple alignments with amino acid sequences from adeno-
sine GPCRs to gather additional support for constructing gene models.
Multiple sequence alignment and phylogenetic analysis. Multiple
sequence alignments were performed by using ClustalX v1.81 (58)
and corrected according to gapped BLASTP results (1). Phylogenetic
analyses were carried out with MEGA v2.1 (38) software, using both
the maximum parsimony and distance-based methods UPGMA (un-
weighted pair group method with arithmetic mean) and neighbor
joining, along with minimum evolution with the Poisson correction
distance method, and gaps were treated by pairwise deletion. Accu-
racy of reconstructed trees was examined by the bootstrap test with
1,000 replications. Phylogenetic trees were rooted with the bovine
rhodopsin sequence. Complete names of organisms included in the
phylogenetic analysis are as follows: ANOGA, Anopheles gambiae
(Arthropoda, insecta); ASTMI, Asterina miniata (starfish; Echinoder-
mata); BOVIN, Bos taurus (Chordata, vertebrata, mammalia);
CAEBR, Caenorhabditis briggsae (Nematoda); CAEEL, Caenorhab-
ditis elegans (Nematoda); CANFA, Canis familiaris (Chordata, ver-
Table 1. Specific agonists and antagonists for ARs used in this work
Abbreviation Chemical Name Receptor Action
Receptor-
Binding Value K
i
Reference
ADSPX 1-allyl-3,7-dimethyl-8-p-sulfophenylxanthine A
2B
Antagonist 0.6 nM (28)
Alloxazine Benzo[g]pteridine 2,4(1H,3H)-dione A
2B
Antagonist 13 nM (40)
CCPA 2-chloro-N
6
-cyclopentyladenosine A
1
Agonist 0.4 nM (43)
CGS-15943 9-chloro-2-(2-furanyl)[1,2,4]triazolo[1,5-
c]quinazoline-5-amine
A
1
Antagonist 4 nM (31)
CGS-21680 2-P(2-carboxyethyl)phenethylamino-5⬘-N-
ethylcarboxamidoadenosine
A
2A
Agonist 15 nM (30)
CSC 1,3,7-trimethyl-8-(3-chlorostyryl) xanthine A
2A
Antagonist 54 nM (29)
DPCPX 8-cyclopentyl-1,3-dipropylxanthine A
1
Antagonist 0.69 nM (25)
IB-MECA 1-deoxy-1-[6-[((3-
iodophenyl)methyl)amino]-9H-purin-9-
yl]-N-methyl-␤-D-ribofuranuronamide
A
3
Agonist 1.1 nM (64)
MRS 1220 9-chloro-2-(2-furanyl)-5-((phenylacetyl)
amino)-[1,2,4]triazol[1,5-c]quinazoline
A
3
Antagonist 14 nM (36)
NECA 5⬘-N-ethylcarboxamidoadenosine A
1
,A
2B
Agonist A
1
⫽11 nM (9)
A
2B
⫽16 nM
AR, adenosine receptor.
E941HOMEOSTATIC ROLE OF INOSINE IN THE LIVER
AJP-Endocrinol Metab •VOL 290 •MAY 2006 •www.ajpendo.org
on March 10, 2009 ajpendo.physiology.orgDownloaded from

tebrata, mammalia); CAVPO, Cavia porcellus (domestic guinea pig;
Chordata, vertebrata, mammalia); CHICK, Gallus gallus (Chordata,
vertebrata, aves); CIOIN, Ciona intestinalis (Chordate, urochordata,
ascidiacea); DANRE, Danio rerio (zebra fish; Chordata, vertebrata,
teleostei); DROME, Drosophila melanogaster (Arthropoda, insecta);
FUGRU, Fugu rubripes (Japanese puffer fish; Chordata, vertebrata,
teleostei); HORSE, Equus caballus (Chordata, vertebrata, mammalia);
HUMAN, Homo sapiens (Chordata, vertebrata, mammalia); MOUSE,
Mus musculus (Chordata, vertebrata, mammalia); RABBIT, Orycto-
lagus cuniculus (Chordata, vertebrate, mammalia); RAT, Rattus nor-
vegicus (Chordata, vertebrata, mammalia); SHEEP, Ovis aries (Chor-
data, vertebrata, mammalia), and XENLA, Xenopus laevis (Chordata,
vertebrata, amphibia).
RESULTS
Inosine stimulates glycogenolysis, gluconeogenesis, and
ureagenesis in hepatocytes via A
3
AR. Adenosine and inosine
concentration-response curves to stimulate glycogenolysis,
gluconeogenesis, and ureagenesis rates are presented in Fig. 1.
Effective concentration (EC
50
) values of adenosine and inosine
were calculated, along with ratios for (adenosine EC
50
value)/
(inosine EC
50
value) in each activated pathway (Table 2).
These data indicated that gluconeogenesis and ureagenesis
might be activated at lower concentrations of inosine than of
adenosine. The stimulating effect of 1 ␮M inosine on glyco-
genolysis, gluconeogenesis, and ureagenesis was blunted spe-
cifically with the selective A
3
AR antagonist 9-chloro-2-(2-
furanyl)-5-[((phenylacetyl)amino)-[1,2,4]triazolo[1,5-c]quina-
zoline (MRS 1220) but was not modified when inosine was
simultaneously incubated with 9-chloro-2-(2-furanyl)[1,2,4]
triazolo[1,5-c]quinazolin-5-amine (CGS-15943), 1,3,7-tri-
methyl-8-(3-chlorostyryl)xanthine (CSC), and 1-allyl-3,7-
dimethyl-8-p-sulfophenylxanthine (ADSPX), or alloxazine, se-
lective antagonists for A
1
,A
2A
, and A
2B
AR, respectively (Fig.
2); i.e, inosine stimulated these three metabolic routes in
isolated rat liver cells only if A
3
AR was not blocked. Two
selective A
2B
AR antagonists were used in these experiments
because the required ADSPX solvent [A
2B
antagonists with
lower receptor-binding constant (Table 1)] is dimethyl sulfox-
ide, which, when used at a concentration of 1 mM to quantify
urea, interfered with the assay (results not shown) (24). Thus,
in this case, ADSPX was substituted for a water-soluble
selective A
2B
AR antagonist such as alloxazine.
Inosine-induced Ca
2
⫹
mobilization to stimulate glycogenol-
ysis, gluconeogenesis, and ureagenesis. A common action of
adenosine and an AR-specific agonist is to increase [Ca
2⫹
]
i
in
isolated hepatocytes (22). Results in Table 3 show that inosine
shares in this action. It is noteworthy that stimulation with
either inosine or the individual AR agonists employed resulted
in a rise in Ca
2⫹
similar to the rise obtained with adenosine,
which might activate all four ARs. We performed three series
of experiments to investigate the role of calcium in liver
metabolic pathway inosine-mediated activation. In the first
series, Ca
2⫹
was eliminated from KRB; in the second series,
EGTA was included in Ca
2⫹
-free KRB to chelate extracellular
Ca
2⫹
; and in the third series, BAPTA-AM was added to
Ca
2⫹
-free KRB to chelate intracellular Ca
2⫹
. Inosine elicited a
lesser stimulation in studied metabolic pathway rates when
cells were incubated in Ca
2⫹
-free KRB. In addition, these
pathways were not stimulated at all by the nucleoside when
either of the used chelating agents was present (Fig. 3).
To identify AR involved in the transient inosine-mediated
increase of free Ca
2⫹
, we conducted the experiment presented
in Fig. 4. Inosine alone produced a temporary increase in Ca
2⫹
(Fig. 4A) that was not modified by A
1
,A
2A
, and A
2B
AR-
selective antagonists (Fig. 4, B–D) but was blunted by A
3
AR
antagonist (Fig. 4E).
Fig. 1. Dose-response curves of inosine (E) or adenosine (F) for glycogenol-
ysis (A), gluconeogenesis (B), and ureagenesis (C) in hepatocytes. Basal values
in the absence of nucleosides (䊐). Plotted values are means, and vertical lines
represent SE of duplicate incubation of 6 independent cell preparations, except
for control sample, where 8 –10 independent cell preparations were included.
Statistical significance vs. control values are indicated. *P⬍0.05; **P⬍
0.001.
Table 2. EC
50
values for adenosine and inosine to stimulate
glycogenolysis, gluconeogenesis, and ureagenesis
in isolated rat hepatocytes
Pathway
EC
50
Values for
Adenosine
EC
50
Values for
Inosine
Ratio, EC
50
Adenosine to
EC
50
Inosine
Glycogenolysis 3.8⫻10
⫺9
M 4.2⫻10
⫺9
M 0.90
Gluconeogenesis 1.7⫻10
⫺8
M 7.8⫻10
⫺9
M 2.2
Ureagenesis 1.8⫻10
⫺7
M 7.0⫻10
⫺8
M 2.6
Data obtained from experiments in Fig. 1. EC
50
, effective concentration.
E942 HOMEOSTATIC ROLE OF INOSINE IN THE LIVER
AJP-Endocrinol Metab •VOL 290 •MAY 2006 •www.ajpendo.org
on March 10, 2009 ajpendo.physiology.orgDownloaded from

Desensitization experiments were conducted to test whether
inosine acted through GPCR. Isolated hepatocytes were stim-
ulated with 1 ␮M adenosine or inosine, and Ca
2⫹
-transient
rises were monitored. After recovery to initial values in ⬃2
min, cells were stimulated again. Under this protocol, adeno-
sine failed to reinitiate cell activation independently of whether
first activation was produced by adenosine (Fig. 5A)orby
inosine (Fig. 5D). Similarly, inosine failed to reinitiate cell
activation independently of whether initial activation was ob-
tained with adenosine (Fig. 5B) or inosine (Fig. 5C).
Inosine- and adenosine-stimulated phosphorylase activity.
Incubation of isolated hepatocytes with either inosine or adeno-
sine resulted in a statistically significant increase in glycogen
phosphorylase activity (Fig. 6). Phosphorylase activity in in-
osine-stimulated cells reached a nearly twofold increase over
basal levels, and this stimulation was blunted with the A
3
AR
antagonist or the intracellular Ca
2⫹
chelant agent BAPTA-AM
(Fig. 6). At equimolecular doses, adenosine was less potent
than inosine, and BAPTA-AM additionally blunted adenosine
stimulation of phosphorylase (Fig. 6). A
3
AR antagonist partial
inhibitory action on adenosine-mediated stimulation (Fig. 6)
might be explained by the effect of adenosine through activa-
tion of AR other than A
3
.
Inosine and cAMP pool. To investigate cAMP involvement
in the inosine response, hepatocytes were incubated with
graded concentrations of the nucleoside. Changes in the cAMP
pool were compared with those produced by adenosine and
AR-selective agonists. Results show that stimulation of A
2A
AR with the selective agonist 2-P(2-carboxyethyl)phenethyl-
amino-5’-N-ethylcarboxamidoadenosine and A
2B
AR with a
mixture of 5⬘-(N-ethylcarboxamido)adenosine (an A
1
,A
2B
AR
agonist) plus 8-cyclopentyl-1,3-dipropylxanthine (an A
1
-selec-
tive AR antagonist) produced a dose-dependent increase in
cAMP content (Fig. 7). In contrast, stimulation with 2-chloro-
N
6
-cyclopentyladenosine, an A
1
-selective AR agonist, or 1-de-
oxy-1-[6-[((3-iodophenyl)methyl)amino]-9H-purin-9-yl]-N-
methyl-␤-D-ribofuranuronamide (IB-MECA), an A
3
AR-
selective agonist, originated a dose-related decrease in cAMP
content (Fig. 7). According to all of the findings in this paper,
inosine actions resemble those of A
3
AR agonists. Therefore, it
would be expected that inosine might decrease cAMP in the
same manner as A
3
AR agonists; however, adenosine and,
unexpectedly, inosine were unable to modify the cAMP pool in
Fig. 2. Effect of inosine in the absence or presence of adenosine receptor
(AR)-selective antagonists on the rate of glycogenolysis (A), gluconeogenesis
(B), and ureagenesis (C) in hepatocytes. Cells were incubated as detailed in
MATERIALS AND METHODS with 1 ␮M inosine alone or combined with 1 ␮M
final concentration of the following AR-selective antagonists: 9-chloro-2-(2-
furanyl)[1,2,4]triazolo[1,5-c]quinazolin-5-amine (CGS-15943) for A
1
; 1,3,7-
trimethyl-8-(3-chlorostyryl)xanthine (CSC) for A
2A
; 1-allyl-3,7-dimethyl-8-p-
sulfophenylxanthine (ADSPX) for for A
2B
; and 9-chloro-2-(2-furanyl)-5-
((phenylacetyl)amino)-[1,2,4]triazolo[1,5-c]quinazoline (MRS 1220) for A
3
.
Control samples were incubated without added inosine or antagonist. In
ureagenesis studies, alloxazine was used instead of ADSPX as an A
2B
-
selective antagonist (see text). Values represent means ⫾SE of duplicate
incubation from 4 to 6 independent cell preparations. *Statistical significance
vs. control sample without inosine, P⬍0.001; **statistical significance
inosine alone vs. inosine ⫹MRS 1220, P⬍0.01.
Table 3. [Ca
2
⫹
]
i
in isolated hepatocytes treated with
adenosine, inosine, or selective AR agonists
Additions AR-Stimulated [Ca
2⫹
]
i
, nmol/l Values, %
None 195⫾6.3 100
Adenosine All 4 284⫾7.3 146
Inosine ? 274⫾6.3 141
CCPA A
1
301⫾6.9 154
CGS-21680 A
2A
281⫾6.9 144
NECA plus DPCPX A
2B
298⫾7.9 153
IB-MECA A
3
279⫾7.9 143
Numbers are means ⫾SE of duplicates from 4 independent cell prepara-
tions. [Ca
2⫹
]
i
, cytosolic Ca
2⫹
concentration. Experimental conditions as in
MATERIALS AND METHODS. To stimulate A
2B
AR alone, an AR agonist for A
2B
and A
1
, such as NECA (Table 1), was mixed with DPCPX, a selective
antagonist for A
1
AR. Nucleosides, agonists, and antagonists were used at a
1-␮M final concentration. Statistical significance, nucleoside or agonist vs.
control; P⬍0.001 in all cases.
E943HOMEOSTATIC ROLE OF INOSINE IN THE LIVER
AJP-Endocrinol Metab •VOL 290 •MAY 2006 •www.ajpendo.org
on March 10, 2009 ajpendo.physiology.orgDownloaded from

hepatocytes (Fig. 7). Next, hepatocytes were incubated with
AR antagonist-supplemented inosine. Thus selective antago-
nists for each of the four ARs, CGS-15943 for A
1
, CSC for
A
2A
, ADSPX for A
2B
, and MRS 1220 for A
3
, were used so that
different mixtures of three of these antagonists added to hepa-
tocytes would maintain three of the four ARs blocked, leaving
only one AR able to be activated, which might or might not be
stimulated by inosine. Only in experiments in which A
3
AR
was not antagonized by the adequate mixture of AR agents did
inosine decrease the cAMP cellular pool (Fig. 8), similarly to
IB-MECA, an A
3
AR agonist, whereas, if inosine was added to
cells in which A
1
,A
2A
,orA
2B
AR were not antagonized by
adequate AR blocker mixtures, cAMP values remained unmodi-
fied (Fig. 8). Additional experiments are required to understand
why inosine alone did not modify the cAMP cellular pool (Fig. 7),
whereas inosine did indeed decrease the cAMP pool if A
1
,A
2A
,
and A
2B
AR were blocked by their selective antagonists (Fig. 8).
Fig. 3. Calcium participation in inosine-mediated stimulation of glycogenolysis (A),
gluconeogenesis (B), and ureagenesis (C) in hepatocytes. Cells were placed under 4
different conditions: 1) complete Krebs-Ringer buffer (KRB) containing 1.2 mM
Ca
2⫹
;2)Ca
2⫹
-free KRB; 3) cells were preincubated for 15 min in Ca
2⫹
-free
KRB supplemented with 1.2 mM EGTA; and 4) cells were preincubated for 20 min
in Ca
2⫹
-free KRB supplemented with 10 ␮M BAPTA-AM. Control cells at left
(filled bars) of each experimental condition and hepatocytes were supplemented with
1⫻10
⫺6
M inosine at the right (open bars) of each experimental condition. Each
datum in the figure corresponds to mean ⫾SE of duplicate incubations from 4 to 6
independent cell preparations. *Statistical significance for cells incubated with 1) KRB
with Ca
2⫹
⫹inosine vs. 2)Ca
2⫹
-free KRB ⫹inosine, P⬍0.01; **3)Ca
2⫹
-free
KRB with EGTA ⫹inosine and 4)Ca
2⫹
-free-KRB with BAPTA-AM ⫹inosine,
both P⬍0.001.
Fig. 4. Effect of inosine in the absence or presence of AR-selective antagonists
on cytosolic Ca
2⫹
concentration ([Ca
2⫹
]
i
) in hepatocytes. Cells were labeled
with fura 2-AM and stimulated with 10
⫺6
M inosine alone or supplemented
with 10
⫺6
M for each AR-selective antagonist indicated. Experiment was
repeated 3 times with identical results.
E944 HOMEOSTATIC ROLE OF INOSINE IN THE LIVER
AJP-Endocrinol Metab •VOL 290 •MAY 2006 •www.ajpendo.org
on March 10, 2009 ajpendo.physiology.orgDownloaded from

In any event, cAMP does not appear to be involved in inosine-
activated metabolic routes in hepatocytes.
Phylogenetic analyses ruled out the existence of an addi-
tional GPCR homologous to ARs in mammals. The results in
this paper, as well as those of other authors, clearly demon-
strate that GPCR, mainly through A
3
AR, mediates some
inosine effects. However, this does not discard the possibility
that other adenosine-related GPCRs might exist, including a
cognate inosine GPCR. To explore the latter possibility, we
conducted an extensive search for homologous protein se-
quences to the adenosine/inosine receptor in whole genomic
DNA sequences from human, mouse, rat, zebra fish (D. rerio),
Japanese puffer fish (F. rubripes), and the ascidian C. intesti-
nalis. Subsequently, we conducted a phylogenetic analysis for
retrieved adenosine/inosine receptor sequences. We found no
additional cognate adenosine/inosine receptors in addition to
the four known adenosine GPCRs in human, mouse, and rat.
Unexpectedly, however, we did find three additional AR-
homologous protein sequences in puffer fish and one in zebra
fish. Recently, a similar observation was reported with ␣
2
-
adrenoceptors, because the zebra fish possesses five ␣
2
-adre-
noceptors instead of the three found in mammals and the puffer
fish possesses eight ␣
2
-adrenoceptors (11, 52, 53). Figure 9
shows a phylogenetic tree constructed with a total of 46
full-length protein sequences identified as ARs (all belonging
exclusively to animals). It can be observed that all AR protein
sequences in mammals belong to one of the four known AR
types. No additional AR types were found in mammals; how-
ever, in the puffer fish, two distinct A
1
AR were found
(designated provisionally as A
1A
and A
1B
) along with one
additional A
2
AR (provisionally denominated A
2C
). On the
other hand, in the complete genome of C. intestinalis (a
nonvertebrate chordate that diverged very early from other
chordates, including vertebrates) we identified only three AR-
homologous protein sequences, although none resulted or-
thologous (same gene in different species) to the four AR types
known in mammals. These three C. intestinalis ARs are
grouped with other AR sequences found in zebra fish, puffer
fish, starfish (echinodermata), arthropoda, and nematoda; these
sequences probably comprise a fifth AR type. Within this
group, only the AR from the starfish A. miniata has been
experimentally demonstrated as an AR coupled to a G
i
-linked
protein (35).
Adenosine, inosine, and glucose are released by the liver
under hypoxia/reoxygenation conditions. Once we defined
which AR was involved in inosine action in liver, the signal
transduction pathway mediating inosine action, and the ab-
sence of an additional adenosine/inosine receptor participating
in these responses, we focused on the physiological meaning of
inosine-mediated action in liver. It is known that adenosine and
inosine can be released by different organs, e.g., brain (39, 65),
heart (34, 41, 56), eye (50), lung (45), kidney, and liver (51).
Furthermore, release of these nucleosides is induced under
hypoxic conditions (34, 41). Isolated rat hepatocytes also
release adenosine under hypoxic conditions (5); however, the
metabolic effect of endogenous adenosine and inosine release
in liver has not been tested. Thereafter, we subjected both
perfused rat liver and isolated hepatocytes to hypoxia/reoxygen-
ation conditions and measured inosine, adenosine, and glucose
release. During hypoxic incubation, isolated hepatocytes accumu-
lated inosine, adenosine, and glucose in extracellular volume
(Table 4). Both nucleosides and glucose accumulation were ob-
served additionally under conditions of hypoxia/reoxygenation.
The selective antagonist for A
3
AR, MRS 1220, impaired libera-
tion of glucose from intracellular sources, but interestingly, it also
impaired inosine and adenosine release from hepatocytes.
We obtained similar results in perfused rat livers that were
subjected to hypoxia and hypoxia/reoxygenation conditions
(Fig. 10). Once experimental conditions were set, inosine,
adenosine, and glucose release began after an initial lag of 2.5
min. Inosine reached a plateau after 10 min and adenosine after
5 min, but glucose increased progressively during the follow-
ing 30 min (Fig. 10).
Fig. 5. Inosine and adenosine desensitize hepatocytes to each other. Cells were
labeled with fura 2-AM and stimulated initially with 10
⫺6
M adenosine (Aand
B)or10
⫺6
M of inosine (Cand D), and free [Ca
2⫹
]
i
was measured as a
function of time. After recovery to initial values, a second stimulation with
inosine (Band C) or adenosine (Aand D) was performed. Experiment was
repeated 4 times with identical results.
E945HOMEOSTATIC ROLE OF INOSINE IN THE LIVER
AJP-Endocrinol Metab •VOL 290 •MAY 2006 •www.ajpendo.org
on March 10, 2009 ajpendo.physiology.orgDownloaded from

Under hypoxia and hypoxia/reoxygenation conditions, glu-
coneogenesis and ureagenesis activities were assayed in rat hepa-
tocytes that were isolated from fasted rats (16 h). Both ATP-
dependent glucose and urea production diminished by 50% in
isolated hepatocytes incubated under hypoxia or hypoxia/reoxy-
genation conditions (data not shown). These latter results can be
explained because under low oxygen tension, insufficient ATP
production precludes flux through anabolic pathways (8).
Fig. 6. Effect of inosine and adenosine of glycogen phosphorylase
activity in hepatocytes from fed rats. Isolated hepatocytes (20 ⫾3mg
of protein) were incubated in 5 ml of Krebs-Ringer bicarbonate
containing 1.2 mM CaCl
2
. Glycogen phosphorylase activity was
measured as detailed in MATERIALS AND METHODS.A: samples were
incubated with 10
⫺6
M inosine alone (F), 10
⫺6
M inosine ⫹10
⫺6
M
MRS 1220 (E), or 10
⫺6
M inosine ⫹10 ␮M BAPTA-AM (). B:
samples were incubated with 10
⫺6
M adenosine (F), 10
⫺6
M adeno-
sine ⫹10
⫺6
M MRS 1220 (E), or 10
⫺6
M adenosine ⫹10 ␮M
BAPTA-AM (). Values are means ⫾SE of 3 independent experi-
ments by duplicate. Statistical significance: P⬍0.001 by comparing
inosine at 0 min vs. inosine at 2.5, 5, and 10 min; P⬍0.001 by
comparing inosine alone vs. inosine ⫹MRS 1220 or inosine ⫹
BAPTA-AM; P⬍0.01 (at least) by comparing adenosine at 0 min vs.
adenosine at 2.5, 5, and 10 min; P⬍0.01 (at least) by comparing
adenosine alone vs. adenosine ⫹MRS 1220 or adenosine ⫹BAPTA-
AM.
Fig. 7. Effect of adenosine, inosine, and AR-selective agonists on cAMP
production in hepatocytes. Cells were incubated for 2 min in KRB with
adenosine (䊐), inosine (F), and the following AR selective agonists: 2-chloro-
N
6
-cyclopentyladenosine for A
1
(); 2-P(2-carboxyethyl)phenethyl-amino-5’-
N-ethylcarboxyamidoadenosine for A
2A
(E), and 1-deoxy-1-[6-[((3-iodophe-
nyl)methyl)amino]-9H-purin-9-yl]-N-methyl-␤-D-ribofuranuronamide for A
3
(ƒ); to stimulate A
2B
AR (■), a mixture of 5’-(N-ethylcarboxamido)adenosine
and 8-cyclopentyl-1,3-dipropylxanthine (DPCPX) was used. Results are ex-
pressed as %basal value, which was 0.74 ⫾0.03 pmol of cAMP formed in 2
min/mg (wet wt). Each value represents means ⫾SE of 4 independent
experiments, each performed in duplicate. Statistical significance vs. basal is
indicated: *P⬍0.05; **P⬍0.01; ***P⬍0.001.
Fig. 8. Effect of inosine on cAMP values of hepatocytes, to which 3 of 4 ARs
were inhibited by mixtures of selective AR antagonists as detailed in the text.
Inosine (1 ␮M, final concentration) was added to each tube in which the AR
noninhibited remnant AR was contained: A
1
() when mixing 1 ␮M (final
concentration) CSC ⫹1␮M ADSPX ⫹1␮M MRS 1220; A
2A
(E) when mixing
1␮M DPCPX ⫹1␮M ADSPX ⫹1␮M MRS 1220; A
2B
(■) when mixing 1 ␮M
DPCPX ⫹1␮M CSC ⫹1␮M MRS 1220; and A
3
(ƒ) when mixing 1 ␮M
DPCPX ⫹1␮M CSC ⫹1␮M ADSPX. Cells were incubated in KRB with the
indicated additions. Results are expressed as %basal value, which was 0.74 ⫾0.03
pmol of cAMP formed in 2 min/mg (wet wt). Each value represents means ⫾SE
of 4 independent experiments, each performed in duplicate. Statistical significance
vs. basal is indicated: *P⬍0.001.
E946 HOMEOSTATIC ROLE OF INOSINE IN THE LIVER
AJP-Endocrinol Metab •VOL 290 •MAY 2006 •www.ajpendo.org
on March 10, 2009 ajpendo.physiology.orgDownloaded from

Fig. 9. Phylogenetic analysis of ARs. Phylogenetic tree constructed with available protein sequences belonging to the AR subfamily by using minimum evolution
method. Trees were calculated using MEGA 2.1 (38). Dotted bars indicate nodes supported in ⬎70 (open), ⬎80 (gray), or ⬎90% (filled) of 1,000 random
bootstrap replicates of all UPGMA (unweighted pair group method with arithmetic mean), neighbor-joining, minimum-evolution, and maximum-parsimony trees.
Scale bar represents 0.2 amino acid substitutions per site. Obtained trees were rooted by use of bovine rhodopsin. Thick vertical bars indicate the taxonomic group
to which the protein sequence belongs and fine vertical bars the type of AR to which the protein sequence belongs. Sequence names are indicated accordingto
a Swiss-Prot-like identifier (gene organism) followed by the database accession number (GenBank, PIR, Swiss-Prot, etc.) and protein amino acid length. AR
sequences deduced from genomic sequences were obtained from the following sources: the Danio rerio Sequencing Group at the Sanger Institute
(http://www.sanger.ac.uk/Projects/D_rerio/), the Fugu rubripes Genome Project v3.0 (2), and the Ciona intestinalis Genome Project v1.0 (16), the last 2 both
at the US Department of Energy Joint Genome Institute (http://genome.jgi-psf.org/fugu6/fugu6.home.html and http://genome.jgi-psf.org/ciona4/
ciona4.home.html). Experimentally characterized ARs are underlined. A full list of organism names included in the tree is provided in MATERIALS AND METHODS.
E947HOMEOSTATIC ROLE OF INOSINE IN THE LIVER
AJP-Endocrinol Metab •VOL 290 •MAY 2006 •www.ajpendo.org
on March 10, 2009 ajpendo.physiology.orgDownloaded from

DISCUSSION
Data from this paper confirm our preliminary finding (23,
68) and greatly extend previous information on the hepatic
actions of inosine. Adenosine- or inosine-stimulated rat liver
cells showed a dose-related increase in the rates of three of the
main hepatic metabolic pathways, namely glycogenolysis, glu-
coneogenesis, and ureagenesis (Table 1 and Fig. 1). Consider-
ing the EC
50
values obtained for adenosine or inosine to
stimulate the three metabolic pathways and the reported phys-
iological concentrations of adenosine and inosine in rat serum
(61), both nucleosides might be effective in producing activa-
tion of these metabolic pathways. Converging evidence sug-
gests a preeminent role of inosine over adenosine in stimulat-
ing hepatic metabolic routes through A
3
AR activation, be-
cause inosine serum concentration is higher than adenosine and
inosine concentration was 25-fold higher than adenosine in
hepatic venous effluents of isolated perfused liver (51). In
addition, EC
50
values for inosine are similar to EC
50
values for
adenosine to stimulate glycogenolysis, but they are 2- to
2.5-fold lower in stimulating gluconeogenesis and ureagenesis.
Furthermore, a very active adenosine deaminase is present in
rat serum that converts adenosine into inosine (48). The pre-
viously mentioned considerations led us to explore some char-
acteristics of inosine-mediated actions on liver cells, although
additional evidence in favor of inosine as an important inter-
cellular messenger will be included in the final part of this
discussion. The three main inosine-stimulated metabolic path-
ways were equally blunted by the selective A
3
AR antagonist
(Fig. 2); in contrast, incubation of liver cells with selective A
1
,
A
2A
,orA
2B
AR antagonists did not modify the inosine-
mediated rise in glycogenolysis, gluconeogenesis, and ure-
agenesis rates (Fig. 2). Hence, A
3
AR seems to be the initial
target of inosine for stimulating metabolic actions in liver cells.
In previous work with isolated hepatocytes (20, 22, 60, 66),
it was established that the change in the [Ca
2⫹
]
i
pool was the
transduction mechanism elicited by A
3
AR stimulation with
adenosine or A
3
AR agonists to obtain glycogenolysis, glu-
coneogenesis, and ureagenesis rate increases. Similar results
were obtained after stimulation of isolated hepatocytes with
inosine: an increase in [Ca
2⫹
]
i
(Table 3), dependence of such
an increase to activate metabolic pathway rates (Fig. 3), and
blockade in the rise of [Ca
2⫹
]
i
observed exclusively with MRS
1220, the A
3
AR antagonist, but not with the use of selective
A
1
,A
2A
, and A
2B
AR antagonists (Fig. 4).
Main metabolic pathway stimulation in liver by inosine is
absolutely dependent on an increase in free [Ca
2⫹
]
i
(Fig. 3).
Thus incubation of cells in Ca
2⫹
-free KRB supplemented with
the intracellular chelant BAPTA-AM impaired any inosine-
mediated activation in glycogenolysis, gluconeogenesis, and
ureagenesis rates (Fig. 3). Nonetheless, when hepatocytes were
incubated in Ca
2⫹
-free KRB in the absence of chelant agents,
inosine produced minor stimulation in the metabolic pathway
rates that we studied compared with stimulation observed in
KRB containing 1.2 mM Ca
2⫹
(Fig. 3). All these data point to
a relevant role of extracellular Ca
2⫹
in inosine-mediated trans-
duction actions in liver and to a minor contribution of intra-
cellular Ca
2⫹
storage compartments to drive the same actions.
Unpublished experiments (Guinzberg R and Pin˜ a E) using
isolated hepatocytes, incubated in KRB with 1.2 mM Ca
2⫹
and
challenged with MRS 1220, an A
3
AR agonist, are confirma-
tory. Thapsigargin, an inhibitor of Ca
2⫹
release from intracel-
lular storage compartments, decreases stimulation of urea syn-
thesis by nearly 40%.
The following experiment presents another property of the
inosine-sensitive AR. This nucleoside desensitizes AR toward
adenosine (Fig. 4); a lower concentration of serum adenosine
will be quantitatively less important compared with inosine to
promote further metabolic responses in liver. In addition, these
data support that a GPCR is involved in inosine-mediated
actions in liver.
Intracellular Ca
2⫹
increase has been shown to stimulate
glycogenolysis (33, 63). In particular, two Ca
2⫹
-mobilizing
agents, namely epinephrine and ionophore A-23187, promoted
hepatocyte glycogen phosphorylase activation that led to an
increase in cell glucose release (62). Thereafter, a [Ca
2⫹
]
i
rise
by A
3
AR stimulation in hepatocytes by any of the studied
nucleosides (Fig. 4) in turn activated glycogen phosphorylase
to a greater extent with inosine than with adenosine (Fig. 6).
With the information recorded to this point in this work, we
could anticipate a blockade in nucleoside-mediated phosphor-
ylase activation with the use of either a selective A
3
AR
antagonist (Fig. 4) or an intracellular chelating agent (Fig. 3).
In fact, both inhibitory actions were recorded (Fig. 6).
Two additional experiments analyzing the role of cAMP as
a signal transduction pathway for inosine-mediated metabolic
actions gave negative results. Inosine alone, as well as adeno-
sine alone, did not modify cAMP pool in liver cells (Fig. 7). In
another set of experiments with isolated hepatocytes (Fig. 8),
Table 4. Release of inosine and adenosine and glycogenolysis rate in isolated rat hepatocytes mantained in Ringer-HEPES
buffer and subjected to different oxygenation conditions
Experimental Conditions
Inosine Adenosine Glycogenolysis
␮mol glucose/g wet
wt in 45 min␮mol䡠min
⫺1
䡠mg prot
⫺1
Oxygenation O
2
-CO
2
(19:1) 2.22⫾0.01 1.86⫾0.02 54.3⫾2.2
O
2
-CO
2
⫹(19:1) 10
⫺6
MRS 1220 1.41⫾0.06* 1.65⫾0.02 50.8⫾1.7
Hypoxia N
2
-CO
2
(19:1) 4.69⫾0.04* 3.89⫾0.01* 123.8⫾3.2*
N
2
-CO
2
(19:1) ⫹10
⫺6
M-MRS 1220 1.48⫾0.01† 1.34⫾0.02† 72.5⫾2.7†
Hypoxia/reoxygenation N
2
-CO
2
for 25 min, then O
2
-CO
2
for 20 min 4.67⫾0.11* 3.03⫾0.01* 125.3⫾2.3*
N
2
-CO
2
⫹10
⫺6
M MRS 1220 for 25 min, then
O
2
-CO
2
⫹10
⫺6
M MRS 1220 for 20 min
2.00⫾0.02‡ 1.68⫾0.01‡ 62.7⫾1.9‡
Values are means ⫾SE in 3 independent experiments with duplicate samples. Hepatocytes were incubated for 45 min in Ringer-HEPES buffer containing
the following: 120 mM NaCl, 1.2 mM CaCl
2
, 1.2 mM MgSO
4
, 1.2 mM KH
2
PO
4
, and 20 mM HEPES (pH7.4). Cell aliquots were incubated at 37°C and gassed
under conditions indicated in the table. Aliquots were withdrawn for analysis at the end of the incubation period. Statistical significance: *P⬍0.01 vs. control
value with O
2
-CO
2
(19:1); †P⬍0.001 vs. value with N
2
-CO
2
(19:1); ‡P⬍0.001 vs. value with N
2
-CO
2
for 25 min, then O
2
-CO
2
for 20 min.
E948 HOMEOSTATIC ROLE OF INOSINE IN THE LIVER
AJP-Endocrinol Metab •VOL 290 •MAY 2006 •www.ajpendo.org
on March 10, 2009 ajpendo.physiology.orgDownloaded from

inosine lowered the cAMP pool and behaved similarly to the
selective A
3
AR agonist IB-MECA (Fig. 7) but only when
selective A
1
,A
2A
, and A
2B
AR antagonists were supplemented
in the incubation mixture (Fig. 8). The significance of these
experiments remains to be evaluated but is inconsistent with
any participation of cAMP in inosine-mediated activation of
metabolic pathways in liver.
Phylogenetic analysis results excluded the existence of ad-
ditional cognate adenosine/inosine receptors in mammals, but
this analysis leads us to propose that the four AR types
observed in mammals, A
1
,A
2A
,A
2B
, and A
3
, arose during the
evolution of early vertebrates. Their origin is related to genome
duplications produced before radiation of jawed vertebrates
some 500 million years ago (26, 55). Phylogenetic analysis
also suggests the probable existence of a fifth type of AR in
invertebrates and lower vertebrates (fishes). This latter finding
agrees with previous papers that claim the presence of adeno-
sine GPCR in nonvertebrate animals such as the blowfly
Calliphora vicina (44), the bloodsucking bug Rhodnius pro-
lixus (12), the mussels Mytilus californianus (13) and Mytilus
edulis (4), and the spiny lobster Panulirus argus (17). Further-
more, one protein within this group (accession no. AAN33001)
has been experimentally demonstrated as an AR in the starfish
A. miniata (35), reinforcing the idea that this group of proteins
probably corresponds to a fifth type of AR. It should be
mentioned that Clark et al. (15), after a great effort to identify
novel human transmembrane proteins, reported an additional
putative AR of 347 amino acid (AA) length (accession no.
AAQ89007). However, this novel protein, predicted from iso-
lated full-length cDNA, is a chimeric protein comprising an
NH
2
-terminal domain identical to the first 119 AA from the A
3
AR and a COOH-terminal domain homologous to single Ig
domain receptor (140 –347 AA) (14). This chimeric protein
results from alternative mRNA splicing, fusing the first exon of
A
3
AR (ADORA3) gene and ADO26 gene located downstream
of ADORA3 gene. However, on the basis of modeling studies
of A
3
AR (46) it can be predicted that the adenosine-binding
domain in this chimeric protein is disrupted and, therefore,
cannot be considered as an AR. In short, the considered
exclusion of additional cognate adenosine/inosine receptors in
humans and rats (Fig. 9) reinforces the previously suggested
central role of hepatic A
3
AR as the physiological receptor for
inosine in preference to adenosine.
Inosine-mediated stimulation of A
3
AR in the liver, through
aCa
2⫹
-dependent process (Figs. 3 and 4E) and independent
from cAMP involvement (Figs. 7 and 8), activates a glycogen
phosphorylase (Fig. 6) and raises glucose release from hepatic
cells (Figs. 1 and 2), an oxidizable substrate that is particularly
useful under ischemic conditions. The final experiments (Table
4 and Fig. 10) link cellular ischemia with inosine/adenosine
release and glucose liberation from liver. For both experimen-
tal designs, isolated hepatocytes and perfused liver, cellular
ischemia produced a nearly twofold increase in inosine/aden-
osine release, slightly higher with inosine than adenosine. The
huge increase in glucose liberation from hepatic glycogen
Fig. 10. Release of inosine, adenosine, and glucose from perfused rat liver
under different oxygenation conditions. All experiments were performed after
a 30-min equilibration period in which liver was perfused with KRB solution
saturated with O
2
-CO
2
mixture (19:1). Under control conditions, the same
perfusion solution (KRB) saturated with an O
2
-CO
2
mixture (19:1) was passed
through the liver for an additional 30 min (F). In hypoxia experiments, the
perfusion solution was replaced with KRB saturated with an N
2
-CO
2
mixture
(19:1) and passed through the liver for additional 30 min (ƒ). In hypoxia/
reoxygenation experiments, the perfusion solution was replaced first with KRB
solution saturated with N
2
-CO
2
mixture (19:1) followed 5 min later with KRB
solution saturated with O
2
-CO
2
mixture (19:1) for 25 min (■). An additional
hypoxia/reoxygenation experiment was performed in identical form, but 10
⫺6
M MRS 1220 was included in the KRB solutions ({). Liver effluent samples
(100 ␮l) were withdrawn at time intervals. Values are means ⫾SE for 3
independent and duplicated experiments. Statistical significance for inosine
values: P⬍0.001 by comparing control vs. the other 3 experimental groups at
all tested times; P⬍0.001 by comparing hypoxia/reoxygenation vs. hypoxia/
reoxygenation ⫹MRS 1220 at all tested times. Statistical significance for
adenosine values: P⬍0.01 by comparing control vs. hypoxia or hypoxia/
reoxygenation; P⬍0.05 by comparing control vs. hypoxia/reoxygenation ⫹
MRS 1220; P⬍0.001 by comparing hypoxia/reoxygenation ⫹MRS 1220 vs.
hypoxia or hypoxia/reoxygenation. Statistical significance for glucose values:
P⬍0.001 by comparing control or hypoxia/reoxygenation ⫹MRS 1220 vs.
hypoxia or hypoxia/reoxygenation at all times tested.
E949HOMEOSTATIC ROLE OF INOSINE IN THE LIVER
AJP-Endocrinol Metab •VOL 290 •MAY 2006 •www.ajpendo.org
on March 10, 2009 ajpendo.physiology.orgDownloaded from

(22.7-fold in Fig. 10) followed the release of nucleosides.
Because this liberation was blunted with the A
3
AR antagonist,
we can conclude that glucose release was due to the presence
of nucleosides. Relevance of the herein-reported studies on
inosine and its physiological role in the liver, stimulating
glucose release, is further supported by the documented pro-
tective role of inosine for a variety of ischemic and inflamma-
tory injuries, particularly in muscular tissues (67). Therefore, it
appears plausible that release of inosine/adenosine under hyp-
oxia conditions in tissues other than liver (34, 41, 50, 51, 56,
61) might promote liberation of glucose from hepatic cells
responding to the activation of both nucleosides, preferably
inosine. In conclusion, we propose that in situations of tissular
ischemia, inosine liberated from different tissues has a physi-
ological role of paramount importance, i.e., to contribute in
maintaining body homeostasis by providing blood glucose
from liver glycogen through A
3
AR activation. In contrast,
although stimulation of specific A
1
,A
2A
, and A
2B
hepatic ARs
resulted in glycogenolysis, gluconeogenesis, and ureagenesis
activation, presently, the effective function of these receptors
in liver cells has not been described.
ACKNOWLEDGMENTS
We are grateful to Adriana Julia´n-Sa´nchez (FM-UNAM) for help with
phylogenetic analyses, Alejandra Palomares for secretarial contribution, and
Ingrid Masher and Maggie Brunner for careful reading of the manuscript.
GRANTS
This work was partially supported by Grant IN211502–2 from Direccion
General de Asuntos del Personal Academico, UNAM, and 45003-A1 from the
Mexican Council of Science and Technology.
REFERENCES
1. Altschul SF and Lipman DJ. Protein database searches for multiple
alignments. Proc Natl Acad Sci USA 87: 5509 –5513, 1990.
2. Aparicio S, Chapman J, Stupka E, Putnam N, Chia JM, Dehal P,
Christoffels A, Rash S, Hoon S, Smit A, Gelpke MD, Roach J, Oh T,
Ho IY, Wong M, Detter C, Verhoef F, Predki P, Tay A, Lucas S,
Richardson P, Smith SF, Clark MS, Edwards YJ, Doggett N,
Zharkikh A, Tavtigian SV, Pruss D, Barnstead M, Evans C, Baden H,
Powell J, Glusman G, Rowen L, Hood L, Tan YH, Elgar G, Hawkins
T, Venkatesh B, Rokhsar D, and Brenner S. Whole-genome shotgun
assembly and analysis of the genome of Fugu rubripes. Science 297:
1301–1310, 2002.
3. Bairoch A and Apweiler R. The SWISS-PROT protein sequence data-
base and its supplement TrEMBL in 2000. Nucleic Acids Res 28: 45– 48,
2000.
4. Barraco RA and Stefano GB. Pharmacological evidence for the modu-
lation of monoamine release by adenosine in the invertebrate nervous
system. J Neurochem 54: 2002–2006, 1990.
5. Belloni FL, Elkin PL, and Giannotto B. The mechanism of adenosine
release from hypoxic rat liver cells. Br J Pharmacol 85: 441– 446, 1985.
6. Benson DA, Karsch-Mizrachi I, Lipman DJ, Ostell J, Rapp BA, and
Wheeler DL. GenBank. Nucleic Acids Res 28: 15–18, 2000.
7. Berry MN and Friend DS. High-yield preparation of isolated rat liver
parenchymal cells. J Cell Biol 43: 506 –520, 1969.
8. Boon L and Meijer AJ. Oxygen tension does not affect urea synthesis in
perfused rat hepatocytes. Eur J Biochem 195: 455–457, 1991.
9. Bruns RF, Lu GH, and Pugsley TA. Characterization of the A2 aden-
osine receptor labeled by [3H]NECA in rat striatal membranes. Mol
Pharmacol 29: 331–346, 1986.
10. Burge C and Karlin S. Prediction of complete gene structures in human
genomic DNA. J Mol Biol 268: 78 –94, 1997.
11. Bylund DB. Alpha-2 adrenoceptor subtypes: are more better? Br J
Pharmacol 144: 159 –160, 2005.
12. Caruso-Neves C, Monteiro SO, de Oliveira CF, Filho CC, and Lopes
AG. Adenosine modulates the (Na(⫹)⫹K(⫹))ATPase activity in mal-
pighian tubules isolated from Rhodnius prolixus. Arch Insect Biochem
Physiol 43: 72–77, 2000.
13. Chen JH and Bayne CJ. Hemocyte adhesion in the California mussel
(Mytilus californianus): regulation by adenosine. Biochim Biophys Acta
1268: 178 –184, 1995.
14. Chung DH, Humphrey MB, Nakamura MC, Ginzinger DG, Seaman
WE, and Daws MR. CMRF-35-like molecule-1, a novel mouse myeloid
receptor, can inhibit osteoclast formation. J Immunol 171: 6541– 6548,
2003.
15. Clark HF, Gurney AL, Abaya E, Baker K, Baldwin D, Brush J, Chen
J, Chow B, Chui C, Crowley C, Currell B, Deuel B, Dowd P, Eaton D,
Foster J, Grimaldi C, Gu Q, Hass PE, Heldens S, Huang A, Kim HS,
Klimowski L, Jin Y, Johnson S, Lee J, Lewis L, Liao D, Mark M,
Robbie E, Sanchez C, Schoenfeld J, Seshagiri S, Simmons L, Singh J,
Smith V, Stinson J, Vagts A, Vandlen R, Watanabe C, Wieand D,
Woods K, Xie MH, Yansura D, Yi S, Yu G, Yuan J, Zhang M, Zhang
Z, Goddard A, Wood WI, Godowski P, and Gray A. The secreted
protein discovery initiative (SPDI), a large-scale effort to identify novel
human secreted and transmembrane proteins: a bioinformatics assessment.
Genome Res 13: 2265–2270, 2003.
16. Dehal P, Satou Y, Campbell RK, Chapman J, Degnan B, De Tomaso
A, Davidson B, Di Gregorio A, Gelpke M, Goodstein DM, Harafuji N,
Hastings KE, Ho I, Hotta K, Huang W, Kawashima T, Lemaire P,
Martinez D, Meinertzhagen IA, Necula S, Nonaka M, Putnam N, Rash
S, Saiga H, Satake M, Terry A, Yamada L, Wang HG, Awazu S,
Azumi K, Boore J, Branno M, Chin-Bow S, DeSantis R, Doyle S,
Francino P, Keys DN, Haga S, Hayashi H, Hino K, Imai KS, Inaba K,
Kano S, Kobayashi K, Kobayashi M, Lee BI, Makabe KW, Manohar
C, Matassi G, Medina M, Mochizuki Y, Mount S, Morishita T, Miura
S, Nakayama A, Nishizaka S, Nomoto H, Ohta F, Oishi K, Rigoutsos
I, Sano M, Sasaki A, Sasakura Y, Shoguchi E, Shin-i T, Spagnuolo A,
Stainier D, Suzuki MM, Tassy O, Takatori N, Tokuoka M, Yagi K,
Yoshizaki F, Wada S, Zhang C, Hyatt PD, Larimer F, Detter C,
Doggett N, Glavina T, Hawkins T, Richardson P, Lucas S, Kohara Y,
Levine M, Satoh N, and Rokhsar DS. The draft genome of Ciona
intestinalis: insights into chordate and vertebrate origins. Science 298:
2157–2167, 2002.
17. Derby CD, Ache BW, and Carr WE. Purinergic modulation in the brain
of the spiny lobster. Brain Res 421: 57– 64, 1987.
18. Fales FW. Glucose (enzymatic). Stand Methods Clin Chem 4: 101–112,
1963.
19. Go´ mez G and Sitkovky MV. Differential requirement for A
2A
and A
3
adenosine receptors for the protective effect of inosine in vivo. Blood 102:
4472– 4478, 2003.
20. Gonza´ lez-Benı´tez E, Guinzberg R, Dı´az-Cruz A, and Pin˜a E. Regula-
tion of glycogen metabolism in hepatocytes through adenosine receptors.
Role of Ca
2⫹
and cAMP. Eur J Pharmacol 437: 105–111, 2002.
21. Guinzberg R, Dı´az-Cruz A, Uribe S, and Pin˜a E. Inhibition of adeno-
sine mediated responses in isolated hepatocytes by depolarizing concen-
trations of K⫹.Biochem Biophys Res Commun 197: 229–234, 1993.
22. Guinzberg R, Dı´az-Cruz A, Uribe S, and Pin˜a E. Ca
2⫹
dependence of
the response of three adenosine type receptors in rat hepatocytes. Eur
J Pharmacol 340: 243–247, 1997.
23. Guinzberg RP, Laguna I, Zentella A, Guzman R, and Pin˜a E.Effect of
adenosine and inosine on ureagenesis in hepatocytes. Biochem J 245:
371–374, 1987.
24. Gutman I and Bergmeyer HU. Determination of urea. In: Methods of
Enzymatic Analysis. New York: Academic, 1974, p. 1791–1794.
25. Haleen SJ, Steffen RP, and Hamilton HW. PD 116,948, a highly
selective A
1
adenosine receptor antagonist. Life Sci 40: 555–561, 1987.
26. Holland PW, Garcı´a-Ferna´ndez J, Williams NA, and Sidow A. Gene
duplications and the origins of vertebrate development. Dev Suppl 125–
133, 1994.
27. Idzko M, Panther E, Bremer HC, Windisch W, Sorichter S, Herouy Y,
Elsner P, Mockenhaupt T, Girolomoni G, and Norgauer J. Inosine
stimulates chemostaxis, Ca2⫹-transients and actin polymerization in im-
mature human dendritic cells via a pertussis toxin-sensitive mechanism
independent of adenosine receptors. J Cell Physiol 199: 149–156, 2004.
28. Jacobson KA, Nikodijevic O, Pedgett WL, Gallo-Rodrı´guez C, Mill-
ard M, and Daly JW. 8-(3-Chlorostyryl)caffeine (CSC) is a selective
A
2
-adenosine antagonist in vitro and in vivo. FEBS Lett 323: 141–144,
1993.
29. Jacobson KA, Shi D, Gallo-Rodrı´guez C, Manning M Jr, Muller C,
Daly JW, Neumeyer JI, Kiriasis I, and Pfleiderer W. Effect of triflu-
E950 HOMEOSTATIC ROLE OF INOSINE IN THE LIVER
AJP-Endocrinol Metab •VOL 290 •MAY 2006 •www.ajpendo.org
on March 10, 2009 ajpendo.physiology.orgDownloaded from

oromethyl and other substituents on activity of xanthines at adenosine
receptors. J Med Chem 36: 2639 –2644, 1993.
30. Jarvis MF, Schulz R, Hutchinson AJ, Do UH, Sills MA, and Williams
M. [3H]CGS 21680, a selective A2 adenosine receptor agonist directly
labels A2 receptors in rat brain. J Pharmacol Exp Ther 251: 888 – 893,
1989.
31. Jarvis MF, Williams M, Do UH, and Sills MA. Characterization of the
binding of a novel nonxanthine adenosine antagonist radioligand,
[3H]CGS 15943, to multiple affinity states of the adenosine A1 receptor in
the rat cortex. Mol Pharmacol 39: 49 –51, 1990.
32. Jin X, Shepherd RK, Duling BR, and Linden J. Inosine binds to A3
adenosine receptors and stimulates mast cell degranulation. J Clin Invest
100: 2849 –2857, 1997.
33. Joseph SK and Williamson JR. The origin, quantitation, and kinetics of
intracellular calcium mobilization by vasopressin and phenylephrine in
hepatocytes. J Biol Chem 258: 10425–10432, 1983.
34. Kekesi V, Zima E, Barat E, Huszar E, Nagy A, Losonczi L, Merkely
B, Horkay F, and Juhasz-Nagy A. Pericardial concentrations of adeno-
sine, inosine and hypoxanthine in an experimental canine model of spastic
ischaemia. Clin Sci (Lond) 48: 198S–201S, 2002.
35. Kalinowski RR, Jaffe LA, Foltz KR, and Giusti AF. A receptor linked
to a Gi-family G-protein functions in initiating oocyte maturation in
starfish but not frogs. Dev Biol 253: 139 –149, 2003.
36. Kim YC, Ji XD, and Jacobson KA. Derivatives of the triazoloquinazo-
line adenosine antagonist (CGS15943) are selective for the human A3
receptor subtype. J Med Chem 39: 4142– 4148, 1996.
37. Krogh A. Two methods for improving performance of an HMM and their
application for gene finding. Proc Int Conf Intell Syst Mol Biol 5:
179 –186, 1997.
38. Kumar S, Tamura K, Jakobsen IB, and Nei M. MEGA2: molecular
evolutionary genetics analysis software. Bioinformatics 17: 1244 –1245,
2001.
39. Latini S and Pedate F. Adenosine in the central nervous system: release
mechanisms and extracellular concentrations. J Neurochem 79: 463– 484,
2001.
40. Liang BT and Haltiwanger B. Adenosine A2a and A2b receptors in
cultured fetal chick heart cells. High- and low-affinity coupling to stimu-
lation of myocyte contractility and cAMP accumulation. Circ Res 76:
242–251, 1995.
41. Linden J. Molecular approach to adenosine receptors: receptor-mediated
mechanisms of tissue protection. Annu Rev Pharmacol Toxicol 41: 775–
787, 2001.
42. Llopis J, Kass GE, Gahm A, and Orrenius S. Evidence for two
pathways of receptor-mediated Ca
2⫹
entry in hepatocytes. Biochem J 384:
243–247, 1992.
43. Lohse MJ, Klotz KN, Schwabe U, CristaLLi G, Vittori S, and Gri-
fantini M. 2-Chloro-N6-cyclopentyladenosine: a highly selective agonist
at A1 adenosine receptors. Naunyn Schmiedebergs Arch Pharmacol 33:
687– 689, 1988.
44. Magazanik LG and Fedorova IM. Modulatory role of adenosine recep-
tors in insect motor nerve terminals. Neurochem Res 28: 617– 624, 2003.
45. Mentzer RM Jr, Rubio R, and Berne RM. Release of adenosine by
hypoxic canine lung tissue and its possible role in pulmonary circulation.
Am J Physiol 229: 1625–1631, 1975.
46. Moro S, Spalluto G, and Jacobson KA. Techniques: recent develop-
ments in computer-aided engineering of GPCR ligands using the human
adenosine A3 receptor as an example. Trends Pharmacol Sci 26: 44 –51,
2005.
47. Olsson RA. Changes in content of purine nucleoside in canine myocar-
dium during coronary occlusion. Circ Res 26: 301–306, 1970.
48. Plagemann PG, Wohlhueter RM, and Kraupp M. Adenosine uptake,
transport, and metabolism in human erythrocytes. J Cell Physiol 125:
330 –336, 1985.
49. Ralevic V and Burnstock G. Receptors for purines and pyrimidines.
Pharmacol Rev 50: 413– 492, 1998.
50. Roth S, Rosenbaum PS, Osinski J, Park SS, Toledano AY, Li B, and
Moshfeghi AA. Ischemia induces significant changes in purine nucleoside
concentration in the retina-choroid in rats. Exp Eye Res 65: 771–779,
1997.
51. Rubio R and Berne RM. Localization of purine and pyridine nucleoside
phosphorylases in heart, kidney, and liver. Am J Physiol Heart Circ
Physiol 239: H721–H730, 1980.
52. Ruuskanen JO, Laurila J, Xhaard H, Rantanen VV, Vuoriluoto K,
Wurster S, Marjamaki A, Vainio M, Johnson MS, and Scheinin M.
Conserved structural, pharmacological and functional properties among
the three human and five zebra fish alpha2-adrenoceptors. Br J Pharmacol
144: 165–177, 2005.
53. Ruuskanen JO, Xhaard H, Marjamaki A, Salaneck E, Salminen T,
Yan YL, Postlethwait JH, Johnson MS, Larhammar D, and Scheinin
M. Identification of duplicated fourth alpha2-adrenergic receptor subtype
by cloning and mapping of five receptor genes in zebra fish. Mol Biol Evol
21: 14 –28, 2004.
54. Shah SP, McVicker GP, Mackworth AK, Rogic S, and Ouellette BF.
GeneComber: combining outputs of gene prediction programs for im-
proved results. Bioinformatics 19: 1296 –1297, 2003.
55. Sidow A. Gen(om)e duplications in the evolution of early vertebrates.
Curr Opin Genet Dev 6: 715–722, 1996.
56. Silva PH, Dillon D, and Van Wylen DG. Adenosine deaminase inhibi-
tion augments interstitial adenosine but does not attenuate myocardial
infarction. Cardiovasc Res 29: 616 – 623, 1995.
57. Starke PE, Hoek JB, and Farber JL. Calcium-dependent and calcium-
independent mechanism or irreversible cell injury in cultured hepatocytes.
J Biol Chem 261: 3006 –3012, 1986.
58. Thompson JD, Gibson TJ, Plewniak F, Jeanmougin F, and Higgins
DG. The CLUSTAL_X windows interface: flexible strategies for multiple
sequence alignment aided by quality analysis tools. Nucleic Acids Res 25:
4876 – 4882, 1997.
59. Tilley SL, Wagoner VA, Salvatore CA, Jacobson MA, and Koller BH.
Adenosine and inosine increase cutaneous vasopermeability by activating
A(3) receptors on mast cells. J Clin Invest 105: 361–367, 2000.
60. Tinton SA, Chow SC, Buc-Caldero´ n PM, and Kass GE. Adenosine
stimulates calcium influx in isolated rat hepatocytes. Eur J Biochem 238:
576 –581, 1996.
61. Traut TW. Physiological concentrations of purines and pyrimidines. Mol
Cell Biochem 140: 1–22, 1994.
62. Villalobos-Molina R and Devlin TM. Effects of tri-Calciphor (trimmer
of 16,16-dimethyl-15-dehydroprostaglandin B1) on glucose metabolism in
liver cells. Biochem Biophys Res Commun 201: 1457–1463, 1994.
63. Villalobos-Molina R, Saavedra-Molina A, and Devlin TM. Effect of
hypoxia and reoxygenation on metabolic pathways in rat hepatocytes.
Arch Med Res 29: 219 –223, 1998.
64. Von Lubitz DK, Carter MF, Deutsch SI, Lin RC, Mastropaolo J,
Meshulam Y, and Jacobson KA. The effects of adenosine A3 receptor
stimulation on seizures in mice. Eur J Pharmacol 275: 23–29, 1995.
65. Winn HR, Rubio R, and Berne RM. Brain adenosine concentration
during hypoxia in rat. Am J Physiol Heart Circ Physiol 241: H235–H242,
1981.
66. Yasuda N, Inove T, Horizoe T, Nagata K, Minami H, Kawata T,
Hoshino Y, Harada H, Yoshikawa S, Asano O, Nagaoka J, Murakami
M, Abe S, Kobayashi S, and Tanaka I. Funtional characterization of the
adenosine receptor contributing to glycogenolysis and gluconeogenesis in
rat hepatocytes. Eur J Pharmacol 459: 159 –166, 2003.
67. Wakai A, Winter DC, Street JT, O’Sullivan RG, Wang JH, and
Redmond HP. Inosine attenuates tourniquet-induced skeletal muscle
reperfusion injury. J Surg Res 99: 311–315, 2001.
68. Zentella de Pin˜ a M, Dı´az-Cruz A, Guinzberg PR, and Pin˜a E. Hor-
mone-like effect of adenosine and inosine gluconeogenesis from lactate in
isolated hepatocytes. Life Sci 44: 2269 –2274, 1989.
E951HOMEOSTATIC ROLE OF INOSINE IN THE LIVER
AJP-Endocrinol Metab •VOL 290 •MAY 2006 •www.ajpendo.org
on March 10, 2009 ajpendo.physiology.orgDownloaded from